U.S. patent application number 10/772278 was filed with the patent office on 2004-08-12 for method of teaching traveling path to robot and robot having function of learning traveling path.
This patent application is currently assigned to Matsushita Electric Industrial Co., Ltd.. Invention is credited to Anezaki, Takashi, Okamoto, Tamao.
Application Number | 20040158358 10/772278 |
Document ID | / |
Family ID | 32820828 |
Filed Date | 2004-08-12 |
United States Patent
Application |
20040158358 |
Kind Code |
A1 |
Anezaki, Takashi ; et
al. |
August 12, 2004 |
Method of teaching traveling path to robot and robot having
function of learning traveling path
Abstract
In a method of teaching a traveling path to a robot, when a
self-propelled robot learns a traveling path, automatic processing
is performed as follows: an instructor only follows the traveling
path, and the self-propelled robot set at a learning mode follows
the traveling path of the instructor and determines path teaching
data. Thus, it is possible to teach a path to the self-propelled
robot without the necessity for the instructor to directly edit
position data.
Inventors: |
Anezaki, Takashi;
(Hirakata-shi, JP) ; Okamoto, Tamao;
(Nishinomiya-shi, JP) |
Correspondence
Address: |
PARKHURST & WENDEL, L.L.P.
1421 PRINCE STREET
SUITE 210
ALEXANDRIA
VA
22314-2805
US
|
Assignee: |
Matsushita Electric Industrial Co.,
Ltd.
Osaka
Kadoma-shi
JP
571-8501
|
Family ID: |
32820828 |
Appl. No.: |
10/772278 |
Filed: |
February 6, 2004 |
Current U.S.
Class: |
700/264 ;
700/245 |
Current CPC
Class: |
G05D 1/028 20130101;
G05D 1/0272 20130101; G05D 2201/0217 20130101; G05D 1/0221
20130101 |
Class at
Publication: |
700/264 ;
700/245 |
International
Class: |
G06F 019/00 |
Foreign Application Data
Date |
Code |
Application Number |
Feb 6, 2003 |
JP |
2003-028949 |
Claims
What is claimed is:
1. A method of teaching a traveling path to a robot, wherein in
teaching a traveling path to an autonomously traveling robot, a
teaching object moves, the robot monitors a position of the
teaching object in time series and detects a movement of the
teaching object based on data on time-series positional changes,
and the robot moves according to the data on positional changes of
the teaching object, and the robot detects a traveling direction
and travel distance of the robot, accumulates the direction and
distance in time series, and converts the direction and distance
into path teaching data.
2. A method of teaching a traveling path to a robot, wherein in
teaching a traveling path to an autonomously traveling robot, a
teaching object moves, the robot autonomously travels according to
taught path teaching data, the robot monitors a position of the
teaching object in time series, detects a movement of the teaching
object based on data on time-series positional change of the
object, and checks the traveling path of the teaching object, and
the robot moves while correcting the taught path teaching data, and
the robot detects a traveling direction and travel distance of the
robot, accumulates the direction and distance in time series, and
converts the direction and distance into path teaching data.
3. A robot having a function of learning a traveling path,
comprising: a position detecting unit for detecting a position of a
teaching object; a movement detecting unit for monitoring the
position of the teaching object in time series and detecting a
movement of the teaching object based on data on time-series
positional changes; a moving unit for moving the robot according to
the data on positional changes of the teaching object; a movement
detecting unit for detecting a traveling direction and travel
distance of the robot; and a data converting unit for accumulating
the movement in time series and converting the movement into path
teaching data.
4. A robot having a function of learning a traveling path,
comprising: a position detecting unit for detecting a position of a
teaching object; a movement detecting unit for monitoring the
position of the teaching object in time series and detecting a
movement of the teaching object based on data on time-series
positional changes of the object; a moving unit for moving the
robot according to taught path teaching data of the robot; and a
control unit for checking a traveling path of the teaching object,
moving the robot while correcting the taught path teaching data,
learning the traveling path of the teaching object while correcting
the taught path teaching data, and determining the path teaching
data.
5. The robot having a function of learning a traveling path
according to claim 3 or 4, wherein the position detecting unit for
detecting a position of the teaching object detects, by using an
array antenna, a signal of a transmitter carried by the teaching
object, whereby the position of the teaching object is
detected.
6. The robot having a function of learning a traveling path
according to claim 3 or 4, wherein the position detecting unit for
detecting a position of the teaching object takes an image of the
teaching object by using a camera, specifies a teaching object
image in a photographing frame, and detects the position of the
teaching object based on a movement of the teaching object
image.
7. The robot having a function of learning a traveling path
according to claim 3 or 4, wherein the position detecting unit for
detecting a position of the teaching object detects the position of
the teaching object by using a sound source direction detecting
unit comprising a directivity sound input member, a signal
direction detecting section, and a direction confirmation control
section.
8. The robot having a function of learning a traveling path
according to claim 3 or 4, wherein the position detecting unit for
detecting a position of the teaching object detects a direction of
a position where the teaching object contacts the robot, whereby
the position of the teaching object is detected.
Description
FIELD OF THE INVENTION
[0001] The present invention relates to a method of teaching a
traveling path to a self-propelled (autonomously moving) robot and
a robot having the function of learning a traveling path.
BACKGROUND OF THE INVENTION
[0002] Conventionally in the field of navigation systems assisting
the driving of automobiles, the following are known: a measuring
section which stores map data and measures the position of an
automobile at each predetermined time, a control section which sets
a display area on the map based on the position measured by the
measuring section, a processing section which generates a display
signal of the map based on the map data read according to the
display area set by the control section, and a device which
performs control such that the display area on the displayed map is
gradually changed from the previously measured position to the
subsequently measured position according to the control of the
control section.
[0003] As a conventional example of a method of teaching an
operation to a robot, the following method is known: a path
teaching device is provided which teaches, to a path following
device, a path to be followed by the end of an operating tool and
displays an actual teaching state on a path teach window, a posture
teaching device is provided which teaches, to the path following
device, a posture to be followed by the operating tool and displays
an actual teaching state on a posture teach window, an operating
state/shape data accumulating device is provided which stores and
accumulates three-dimensional shape data outputted from a shape
measuring device and robot end position information outputted from
the path following device, an accumulated data inspecting device is
provided which calculates various kinds of attribute information
included in the three-dimensional shape data and the robot end
position information according to the specification of an
instructor and displays the calculation results on a data
inspection window, and thus information about changes in the
attributes of sensor data can be visually provided to the
instructor.
[0004] In such a conventional method of teaching a path to a robot,
the user has to directly edit numerical or visual information to
teach position data. However, considering the promotion of robots
for home use, it is not practical that the user directly edits
numerical or visual information to teach position data when
teaching a traveling path to a robot. Thus, a practical method of
teaching a path is necessary.
[0005] An object of the present invention is to provide a method of
teaching a traveling path to a robot that makes it possible to
teach a path to a robot without the necessity for the user, who
teaches the path, to directly edit position data.
DISCLOSURE OF THE INVENTION
[0006] In a method of teaching a traveling path to a robot
according to the present invention, when a traveling path is taught
to an autonomously traveling robot, a teaching object moves, the
robot monitors the position of the teaching object in time series
and detects the movement of the teaching object based on data on
time-series positional changes of the object, and the robot is
moved according to the data on the position changes of the teaching
object, and the robot detects a traveling direction and travel
distance of the robot, accumulates the direction and distance in
time series, and converts the direction and distance into path
teaching data.
[0007] Also, in a method of teaching a traveling path to a robot
according to the present invention, when a traveling path is taught
to an autonomously traveling robot, a teaching object moves, the
robot autonomously travels according to taught path teaching data,
the robot monitors the position of the teaching object in time
series, detects the movement of the teaching object based on data
on time-series positional changes, and checks the traveling path of
the teaching object, the robot is moved while correcting the taught
path teaching data, and the robot detects a traveling direction and
travel distance of the robot, accumulates the direction and
distance in time series, and converts the direction and distance
into path teaching data.
[0008] A robot having a function of learning a traveling path
according to the present invention, comprises a position detecting
unit for detecting the position of a teaching object, a movement
detecting unit for monitoring the position in time series and
detecting the movement of the teaching object based on data on
time-series positional changes, a moving unit for moving the robot
according to the data on the positional changes of the teaching
object, a movement detecting unit for detecting the traveling
direction and travel distance of the robot, and a data converting
unit for accumulating the movement in time series and converting
the traveling direction and travel distance into path teaching
data.
[0009] Also, a robot having the function of learning a traveling
path according to the present invention, comprises a position
detecting unit for detecting the position of a teaching object, a
movement detecting unit for monitoring the position in time series
and detecting the movement of the teaching object based on data on
time-series positional changes of the object, a moving unit for
moving the robot according to taught path teaching data of the
robot, and a control unit for checking the traveling path of the
teaching object, moving the robot while correcting the taught path
teaching data, learning the traveling path of the teaching object
while correcting the taught path teaching data, and determining
path teaching data.
[0010] Further, the position detecting unit for detecting the
position of the teaching object detects, by using an array antenna,
a signal of a transmitter carried by the teaching object, whereby
the position of the teaching object is detected.
[0011] Further, the position detecting unit for detecting the
position of the teaching object takes an image of the teaching
object by using a camera, specifies a teaching object image in a
photographing frame, and detects the position of the teaching
object based on the movement of the teaching object image.
[0012] Still further, the position detecting unit for detecting the
position of the teaching object detects the position of the
teaching object by using a sound source direction detecting unit
which comprises directivity sound input members, a signal direction
detecting section, and a direction confirmation control
section.
[0013] Still further, the position detecting unit for detecting the
position of the teaching object detects a direction of a position
where the teaching object contacts the robot, whereby the position
of the teaching object is detected.
BRIEF DESCRIPTION OF THE DRAWINGS
[0014] FIG. 1 is a structural diagram showing a specific
self-propelled robot for use in a method of teaching a traveling
path to the robot, according to (Embodiment 1) of the present
invention;
[0015] FIG. 2 is an explanatory view showing the teaching of a path
to follow according to the embodiment;
[0016] FIG. 3 is an explanatory view showing the self-propelled
robot, an instructor, and teaching data according to the
embodiment;
[0017] FIG. 4 is an explanatory diagram showing a principle of
detecting a position according to the embodiment;
[0018] FIG. 5 is an explanatory diagram showing an assumed
following operation;
[0019] FIG. 6 is an explanatory diagram showing that the position
of the instructor is monitored in time series and the movement of
the instructor is detected based on the time-series positional
change data according to the embodiment;
[0020] FIG. 7 is an explanatory view showing that a camera is used
as a position detecting unit, according to (Embodiment 2) of the
present invention;
[0021] FIG. 8 is an explanatory view showing that a robot detects
an instructor moving behind the robot and learns a path, according
to (Embodiment 3) of the present invention;
[0022] FIG. 9 is a structural diagram showing a position detecting
unit according to (Embodiment 4) of the present invention; and
[0023] FIG. 10 is a structural diagram showing a position detecting
unit according to (Embodiment 5) of the present invention.
DESCRIPTION OF THE EMBODIMENTS
[0024] A method of teaching a traveling path to a robot of the
present invention will be described below in accordance with the
following specific embodiments.
[0025] (Embodiment 1)
[0026] FIG. 1 shows the configuration of a self-propelled robot
1.
[0027] The self-propelled robot 1 is a robot which autonomously
travels so as to follow a predetermined traveling path without the
necessity for a magnetic tape or a reflection tape partially
provided on a floor as a guide path.
[0028] A moving unit 10 controls the back-and-forth motion and the
lateral motion of the self-propelled robot 1. The moving unit 10 is
constituted of a left-side motor driving section 11 which drives a
left-side traveling motor 111 to move the self-propelled robot 1 to
the right and a right-side motor driving section 12 which drives a
right-side traveling motor 121 to move the self-propelled robot 1
to the left. Driving wheels (not shown) are attached to the
left-side traveling motor 111 and the right-side traveling motor
121.
[0029] A travel distance detecting unit 20 detects a travel
distance of the self-propelled robot 1 which is moved by the moving
unit 10. The travel distance detecting unit 20 is constituted of a
left-side encoder 21 and a right-side encoder 22. The left-side
encoder 21 generates a pulse signal proportionate to the number of
revolutions of the left-side driving wheel driven by the control of
the moving unit 10, that is the number of revolutions of the
left-side traveling motor 111, and detects a travel distance of the
self-propelled robot 1 which has moved to the right. The right-side
encoder 22 generates a pulse signal proportionate to the number of
revolutions of the right-side driving wheel driven by the control
of the moving unit 10, that is the number of revolutions of the
right-side traveling motor 121, and detects a travel distance of
the self-propelled robot 1 which has moved to the left.
[0030] A control unit 50 for operating the moving unit 10 is mainly
constituted of a microcomputer.
[0031] As shown in FIG. 2, (Embodiment 1) will describe an example
in which the self-propelled robot 1 subjected to teaching learns a
path while following an instructor 700 who moves along a path 100
to be taught. In this example, the instructor 700 who moves along
the path 100 to be taught acts as a teaching object.
[0032] A direction angle detecting unit 30 serves as a position
detecting unit for detecting the position of a teaching object. As
shown in FIGS. 3 and 4, the direction angle detecting unit 30
detects, by using an array antenna 501, a signal 500 of a
transmitter 502 carried by the instructor 700, and detects a change
in the traveling direction of the self-propelled robot 1 driven by
the moving unit 10.
[0033] To be specific, in the pickup of the signal 500, the signal
500 is received by the combination of a receiving circuit 503, an
array antenna control section 505, and a beam pattern control
section 504 while the receiving direction of the array antenna 501
is switched. When the receiving level reaches the maximum received
signal level, a beam pattern direction is detected as the direction
of the transmitter 502. Direction angle information 506 acquired
thus is provided to the control unit 50.
[0034] A movement detecting unit 31 monitors direction angles
detected by the direction angle detecting unit 30 in time series
and detects the movement of the instructor 700 based on data on
time-series direction angles. In (Embodiment 1), the time-series
positions of the instructor who moves ahead are detected as changes
in direction angle.
[0035] A movement detecting unit 32 moves the robot according to
the movement of the instructor 700 based on the detection performed
by the movement detecting unit 31, and detects the traveling
direction and the travel distance of the robot from the travel
distance detecting unit 20.
[0036] A data converting unit 33 accumulates movement data in time
series and converts the data into path teaching data 34.
[0037] In a period during which the traveling path is taught, the
control unit 50 reads the travel distance data detected by the
travel distance detecting unit 20 and the traveling direction data
detected by the direction angle detecting unit 30 at each
predetermined time, calculates the current position of the
self-propelled robot 1, controls the traveling of the
self-propelled robot 1 according to the information results, and
performs operation control so that the self-propelled robot 1
follows the traveling path of the instructor.
[0038] When teaching is completed and the path teaching data 34 is
determined (when learning is completed), the control unit 50
performs operation control as that a target path is followed
according to the path teaching data 34 and traveling is accurately
carried out to a target point without deviating from a normal
track.
[0039] In this way, when the self-propelled robot 1 learns a
traveling path, automatic processing is performed as follows: the
instructor 700 only follows the traveling path, and the
self-propelled robot 1 set at a learning mode follows the traveling
path 100 of the instructor 700 and determines the path teaching
data 34. Thus, it is possible to teach a path to the robot without
the necessity for the instructor 700 to directly edit position
data.
[0040] As shown in FIG. 5, when the self-propelled robot 1 set at
the learning mode follows the traveling path 100 of the instructor
along a direction 101 at the shortest distance, accurate teaching
cannot be performed. With the control section 50, as shown in FIG.
6, the self-propelled robot 1 first stores the directions and
distances of the instructor 700 in a sequential manner, and the
self-propelled robot 1 simultaneously calculates the positions (xy
coordinates) of the instructor based on the directions and the
distances and stores the positions. Then, the self-propelled robot
1 detects the relative positions of the stored position data
string, and calculates change points along the direction of the
path 100 based on time-series positions shown in FIG. 6 instead of
the direction 101 at the shortest distance shown in FIG. 5. The
self-propelled robot 1 determines and stores the change points as a
path to be learned. Thus, the self-propelled robot 1 can
autonomously travel accurately along the traveling path 100 of the
instructor 700.
[0041] (Embodiment 2)
[0042] In (Embodiment 1), the position detecting unit detects, as a
change in azimuth angle, the position of the transmitter 502
carried by the instructor 700 in a state in which the array antenna
501 is mounted in the self-propelled robot 1. (Embodiment 2) is
different only in that a camera 801 is mounted on a self-propelled
robot 1 as shown in FIG. 7 to take an image of an instructor 700
who moves ahead, the image of the instructor 700 (instructor image)
is specified on the taken image, and a change in the position of
the instructor 700 on the image is converted into a direction
angle. Besides, in order to specify a taken image of the instructor
700, the instructor 700 wears, for example, a jacket with a
fluorescent-colored marking.
[0043] In this way, even when the camera 801 is used as a position
detecting unit for detecting the position of the instructor who
moves ahead, a traveling path can be similarly taught to the
self-propelled robot 1.
[0044] (Embodiment 3)
[0045] In the above-described embodiments, the self-propelled robot
1 autonomously travels so as to follow the instructor 700 and
learns teaching data. The configuration of FIG. 8 is also
applicable: the self-propelled robot 1 travels ahead of an
instructor 700 according to taught path teaching data, monitors the
position of the instructor 700, who travels behind, in time series
by using the array antenna of (Embodiment 1) or the camera 801 of
(Embodiment 2), detects the movement of the instructor based on
data on time-series positional changes of the instructor, moves the
self-propelled robot 1 according to the movement of the instructor,
compares the movement of the instructor with the taught path
teaching data to check whether or not the instructor follows the
robot along the traveling path, learns the traveling path of the
instructor and performs automatic processing while correcting the
taught path teaching data, and determines path teaching data
34.
[0046] (Embodiment 4)
[0047] FIG. 9 shows (Embodiment 4) which is different from the
above-described embodiments only in the configuration of a position
detecting unit for detecting the position of a teaching object.
[0048] In this case, a sound source direction detector 1401 serving
as a position detecting unit is mounted on the self-propelled robot
1 which is subjected to teaching. An instructor 700 serving as a
teaching object moves along a traveling path to be taught while
uttering a predetermined teaching phrase (e.g. "come here").
[0049] The sound source direction detector 1401 is constituted of
microphones 1402R and 1402L, each serving as a directivity sound
input member, first and second sound detecting sections 1403R and
1403L, a learning signal direction detecting section 1404 serving
as a signal direction detecting section, and a sound
direction-carriage direction feedback control section 1405 serving
as a direction confirmation control section.
[0050] The microphone 1402R and the microphone 1402L detect ambient
sound and the first sound detecting section 1403R detects only the
sound component of the teaching phrase from the sound detected by
the microphone 1402R. The second sound detecting section 1403L
detects only the sound component of the teaching phrase from the
sound detected by the microphone 1402L.
[0051] The learning signal direction detecting section 1404
performs signal pattern matching in each direction and removes a
phase difference in each direction. Further, the learning signal
direction detecting section 1404 extracts a signal intensity from a
sound matching pattern, adds microphone orientation information,
and performs direction vectorization.
[0052] At this point of time, the learning signal direction
detecting section 1404 performs learning beforehand based on the
basic pattern of a sound source direction and a direction vector
and stores learning data therein. Further, in the case of
insufficient accuracy of detecting a sound source, the learning
signal direction detecting section 1404 finely moves (rotates) the
self-propelled robot 1, detects a direction vector at an
approximate angle, and averages the direction vector, so that
accuracy is improved.
[0053] A carriage 1406 of the self-propelled robot 1 is driven
based on the detection results of the learning signal direction
detecting section 1404 via the sound direction-carriage direction
feedback control section 1405, and the self-propelled robot 1 is
moved in the incoming direction of the teaching phrase uttered by
the instructor. Hence, as with (Embodiment 1), the traveling
direction and the travel distance of the robot are detected from a
traveling distance detecting unit 20, and a data converting unit 33
accumulates movement data in time series and converts the data into
path teaching data 34.
[0054] (Embodiment 5)
[0055] FIG. 10 shows (Embodiment 5) which is different from the
above-described embodiments only in the configuration of a position
detecting unit for detecting the position of a teaching object.
[0056] FIG. 10 shows a touching direction detecting unit 1501 which
is mounted on a self-propelled robot 1 instead of the sound source
direction detecting unit. The touching direction detecting unit
1501 decides a state of a teaching touch performed by an instructor
on the touching direction detecting unit 1501 and detects the
position of the instructor.
[0057] A touching direction sensor 1500 mounted on the
self-propelled robot 1 is constituted of a plurality of strain
gauges, e.g., 1502R and 1502L attached to a deformable body 1500A,
just like a load cell device known as a weight sensor. When an area
1500R of the deformable body 1500A is touched, the strain gauge
1502R detects greater strain than that of the strain gauge 1502L.
When an area 1501L of the deformable body 1500A is touched, the
strain gauge 1502L detects greater strain than the strain gauge
1502R. Besides, at least a part of the deformable body 1500A is
exposed from the body of the self-propelled robot 1.
[0058] In a learning touching direction detecting section 1504,
signals detected by the strain gauges 1502R and 1502L are received
via first and second signal detecting sections 1503R and 1503L and
the input signals are separately subjected to signal pattern
matching to detect a peak signal. Further, a plurality of peak
signal patterns are subjected to matching to perform direction
vectorization.
[0059] The learning touching direction detecting section 1504
learns the basic pattern of a touching direction and a direction
vector beforehand and stores learning data therein.
[0060] A carriage 1506 of the self-propelled robot 1 is driven
based on the detection results of the learning touching direction
detecting section 1504 via a touching direction-carriage direction
feedback control section 1505 to move the self-propelled robot 1
along the direction of a touch made by the instructor on the
deformable body 1500A.
[0061] Hence, as with (Embodiment 1), the traveling direction and
the travel distance of the robot are detected from a traveling
distance detecting unit 20, and a data converting unit 33
accumulates movement data in time series and converts the data into
path teaching data 34.
[0062] In (Embodiment 5), a plurality of strain gauges are attached
to the deformable body 1500A to constitute the touching direction
sensor 1500. A plurality of strain gauges may be attached to the
body of the self-propelled robot 1 to constitute the touching
direction sensor 1500.
[0063] As described above, according to the method of teaching a
traveling path to a robot of the present invention, the robot
learns a traveling path while detecting a teaching object moving
along the traveling path to be taught, and performs automatic
processing to determine path teaching data. Thus, an instructor
does not have to directly edit position data, achieving more
practical teaching of a path as compared with the conventional
art.
[0064] Further, the directivity sound input members, the signal
direction detecting section, and the direction confirmation control
section are provided as the position detecting unit for detecting
the position of a teaching object, and the position of the teaching
object is detected by the sound source direction detecting unit.
Also in this configuration, the robot learns a traveling path while
detecting the teaching object who utters a teaching phrase and
moves along the traveling path to be taught, and the robot performs
automatic processing to determine path teaching data. Thus, an
instructor does not have to directly edit position data, achieving
more practical teaching of a path as compared with the conventional
art.
[0065] Moreover, as the position detecting unit for detecting the
position of a teaching object, a direction of a contact made by a
teaching object on the robot is detected and the position of the
teaching object is detected. In this configuration, the teaching
object only has to touch the moving robot so as to indicate a
direction of approaching a traveling path to be taught, and the
robot detects and learns the teaching path and performs automatic
processing to determine path teaching data. Thus, an instructor
does not have to directly edit position data, achieving more
practical teaching of a path as compared with the conventional
art.
* * * * *